Estimate of Combined MuonCalorimeter Trigger Rates

Available on CMS information server CMS NOTE

1998/075

The Compact Muon Solenoid Experiment

Mailing address: CMS CERN, CH-1211 GENEVA 23, Switzerland

CMS Note

17November1998 Estimate of Combined

Muon/Calorimeter T rigger Rates

J.Pliszka

Institute of Theoretical Physics,Warsaw University,Poland

G.Wrochna

CERN,Geneva,Switzerland

Abstract

Rates of combined muon/calorimeter triggers have been studied.The study was based on PYTHIA simulation package and a collection of subroutines for approximate,but very fast estimation of trigger response.Rates of independent muon and calorimeter triggers have been also calculated for a cross-check with detailed simulation.

As a result a set of trigger thresholds is proposed to maintain the1st level trigger output rate at30kHz.

The thresholds are low enough to ful?l the CMS physics program presented in the CMS Technical Proposal.

Introduction

The CMS2nd Level Trigger is design to receive up to100kHz events[1].The1st Level is assumed to deliver not more than30kHz in order to ensure a safety margin.This bandwidth should be divided between muon and calorimeter triggers.Rates of calorimeter triggers,i.e.one or two electrons or photons;electron from a b-quark decay;1,2,3or4jets;electron/photon+jet;missing transverse energy and total transverse energy,have been already calculated[2]-[12].The muon trigger was also simulated and the rates of one and two muon trigger are known[13]-[17].The aim of the present study is to calculate combined muon/calorimeter trigger rates,namely -e,-jet,-,and-.

The study was done for the old ECAL geometry with towers of66crystals.However,we decided to publish the results because no dramatic changes in trigger rates are expected going to55towers.In any case,more detailed study should be done with more sophisticated simulation tools and the results presented here should be considered only as the?rst approximation.

Combined muon/calorimetric triggers have a great importance for many processes to be studied in the future pp colliders.The presence of a high energetic muon gives a very good event signature mainly due to the signi?cant improvement in the signal/noise ratio.In case of the CMS detector in many processes it might be suf?cient to use muon trigger alone but there are processes where only combined muon/calorimetric signature makes the study feasible.There are also several processes for which the combined signature is expected to improve the ef?ciency signi?cantly since the muon requirement should allow for important reduction in calorimetric cut values extending in this way the physics potential.These processes are enlisted in Table1.

Table1:Physics channels involving combined muon/calorimeter triggers.

physics channel jet

H;W+

b or e

B D KK+

?g?g,?q?q1-4+X+

2-3’s

1Simulation

The main dif?culty of this study comes from a huge statistics needed to calculate trigger rates.In order to accom-plish the study in a limited amount of CPU time several ideas have been exploited.Among them are:

events generation in bins,

repeated fragmentation of the event skeleton,

early rejection of non relevant particles and events,

custom(non-GEANT)particle tracking and decays,

fast(non-GEANT)simulation of the calorimeter response with parametrised showers,

parametrisation of the muon trigger response.

Here we describe them only brie?y and we refer the reader to the software description in Ref.[18].

The software consists of the following parts:

event generation,

particle tracking and decays in the detector,

simulation of the detector response,

simulation of the trigger algorithms,

?nal analysis.

1.1Event generation

Events were generated using PYTHIA5.7.Default parameters were taken where possible.Since the trigger rates are dominated by minimum bias events the main effort was put on the simulation of this process.Hard jets were simulated together with a soft background selecting option MSEL=1.In order to collect enough statistics for hard collisions the events were generated in several bins of parton transverse momentum(see Tab.1.1).In the?nal analysis each bin was weighted with a cross section given by PYTHIA.Separate samples were generated for the combined muon/calorimeter triggers and for the calorimeter triggers.The later one was used to cross-check the results with previous studies.

Table2:Generated samples.Notation“2500010”means that25000event

skeletons were generated and each one was fragmented10times(see text).

(GeV)calo events

5-10GeV2500

2500010

20-50GeV12500

2500010

GeV2500

AND Except the2trigger above GeV,where the dominant source of muons is Z[17].

Due to the CPU time consumption it is very important to reject on the earliest possible stage events which are very unprobable to give a trigger.Since calculation of the possible response of the calorimetric trigger is very time consuming,?rst the simple geometrical acceptance for muons was checked.Only events with muons with over 1GeV and were selected for further processing,namely for simulating the calorimeter trigger response. Obviously this condition was not required for the calorimeter trigger sample.

1.3Calorimeter trigger simulation

1.3.1Calorimeter response

Calorimeter trigger is based on three kind of detectors:electromagnetic calorimeter ECAL(),hadronic calorimeter HCAL(),and very forward calorimeter VFCAL().The present study are done for and therefore the VFCAL is not used.

HCAL readout is arranged in towers of.This size de?nes calorimeter trigger cell. ECAL is made out of PbWO crystals.Each crystal in the barrel has a size of,thus each trigger cell contains crystals(see Fig.1).Each cell is divided into6strips of

i.e.crystals.In the endcaps the number of crystals per cell depends on pseudorapidity.

The calorimeter response was obtained with modi?ed CALTRIG package[19].The energy of each particle entering calorimeters was deposited in the ECAL and HCAL cells.For this purpose a realistic geometry of the cells and parametrisations of electromagnetic and hadronic showers was used.A stochastic noise was added to each cell and the digitisation was applied.It consists of cutting the signals which are below the threshold(=3)and rounding the energy values to multiples of0.5or1.0GeV(see Tab.3).

Table3:Parameters of the calorimeter simulation.

ECAL noise(raw energy)

=0.25GeV per tower

HCAL

ECAL

HCAL

1.3.2Calorimeter

The calorimeter package[20].Modi-?cations,which were summarise simulated trigger algorithms.

Trigger primitives.The following trigger primitives are generated by the calorimeter front-end electronics: transverse energy inside an ECAL cell,

?ne grain local isolation bit LI,

transverse energy inside an HCAL cell,

MIP bit–energy deposit compatible with minimum ionising particle.

The LI bit for each cell is computed in the following way.For each pair of adjacent strips in the ECAL cell a sum of transverse energy deposits is calculated.The largest one is found.The ratio is compared to a programmable threshold.If the LI bit is set.The trigger primitives are used as a basic information for all the calorimeter triggers.

Electron/photon trigger.Let us introduce the following symbols for calorimeter cells and transverse energy deposited in them(see Fig.1):

—the ECAL cell containing most of the energy

—cell with maximal of four neighbours

—the HCAL cell behind the

—sum of of8cells around

—sum of of5cells(L-shaped”corner”)around

—the electron/photon threshold

An electron/photon candidate has to ful?l requirements listed in Tab.4.Different sets of thresholds were used for low and high luminosity,as suggested in Ref.[6].

Table4:Electron/photon trigger cuts.

observable cut for

lateral shower pro?le

hadronic isolation GeV

At least one of four GeV

symbol

lateral shower pro?le

longitudinal shower pro?le

hadronic isolation not used

electromagnetic isolation not used

transverse energy threshold variable

Jet triggers.For the purpose of jet triggers the jet transverse energy was de?ned as a sum of the transverse energies computed in a certain calorimeter region(cells).It was used as a threshold for single-and multijet triggers.In the case of multijet triggers a separation of minimum one cell between two jet candidates was required.

Missing and total trigger.These triggers also use the sum of the transverse energies computed in calorimeter regions(cells).The missing energy is a vector and the total energy is a scalar sum of the values for all regions.and were used as thresholds for the respective triggers.

1.3.3Comparison with detailed simulation

The results of the analysis of the calo sample were compared to the results of the detailed simulation done with the CMSIM package,based on GEANT.They are presented in Figures2-6.

In the case of electron/photon and b-electron triggers the comparison is shown after each cut.The agreement between the fast and detailed simulation after all cuts is satisfactory.The e/rate at low is slightly overestimated which means that our conclusions will be rather conservative.

threshold (GeV)

jet E

Figure4:Jet trigger rate.Histogram—fast simulation(this study),points—detailed simulation[23].

The agreement of jet trigger rate shown in Fig.4is impressive.The missing energy is overestimated as much as factor2.Being a vector sum this quantity is the most sensitive to the geometrical details and one should not expect that the simple simulation can reproduce it well.Our result we can consider,however,as an upper limit of the expected trigger rate.

In the case of the trigger only one point from the detailed simulation was available.The agreement at this

point

missing E t threshold (GeV)

m i s s i n g E t t r i g g e r r a t e (k H z ) a t L =1

033

1010101010Σ E t threshold (GeV)

Σ E t t r i g g e r r a t e (k H z ) a t L =1033

Figure 5:trigger rate.Histogram —fast simula-tion (this study),points —detailed simulation [12].

Figure 6:

trigger rate.Histogram —fast simula-tion (this study),point —detailed simulation [23].

1.4Muon

trigger simulation

1.4.1Simulation software

In order to save CPU time we did not simulated explicitly muon tracks,muon detector response and trigger algo-rithms.Instead we used slightly modi?ed version of the parametrisation EFFMRPC described in [21].It gives the response of the Pattern Comparator Trigger (PACT)which compares each pattern of hit RPC strips to prede?ned patterns corresponding to various .The algorithm is described in details elsewhere [24].

1.4.2Comparison with detailed simulation

One and two muon trigger rates obtained in this study are compared to the results of detailed simulation in Figures 7and 8respectively.The detailed simulation [17]was done with the CMSIM package.The agreement is satisfactory.

muon p t cut (GeV)

μ t r i g g e r r a t e (H z )

0.1.1.1.2.2.3.3.4.5.6.7.8.10.12.14.17.20.25.30.35.40.50.70.100.

muon p t cut (GeV)

2μ t r i g g e r r a t e (H z )

0.1.1.1.2.2.3.3.4.5.6.7.8.10.12.14.17.20.25.30.35.40.50.70.100.Figure 7:

Single muon trigger rate at

.Histogram —fast simulation (this

study),points —detailed simulation [17].

Figure 8:

Two muon trigger rate at

.Histogram —fast simulation (this

study),points —detailed simulation [17].

2Results

Obtained trigger rates are presented in Tables 6-12.Each table cell contains the trigger rate (in 100Hz units)for objects with or above the given thresholds.

Table 6:Two muon trigger rate in 100Hz units at

as a function of

cuts on the two muons.

muon p t (GeV) cut

2n d m u o n p t (G e V )

44444444321144444443211444444321144444321144443211333211221111110.0

1.01.21.5

2.02.5

3.03.5

4.0

5.0

6.0

7.0

8.010.012.014.017.020.025.030.035.040.050.070.0100.00.0

1.0

1.2

1.5

2.0

2.5

3.0

3.5

4.0

5.0

6.0

7.0

8.0

10.0

12.0

14.0

17.0

20.0

25.0

30.0

35.0

40.0

50.0

70.0

100.0

Table 7:Cumulative single-and two-muon trigger rate in 100Hz units at as a function of the

cuts.The two-muon cut is the same for both muons.

single muon p t cut (GeV) cut

d o u b l

e m u o n p t c u t (G e V ) c u t

725

697

620

493

374

294

169

100

7257257257256976204933742941691007245282015129877666672572572572569762049337429416910072452820151298776666725725725725697620493374294169100724528201512987766667257257257256976204933742941691007245282015129876666672572572572569762049337429416910072452720151298766655725725725725697620493374294169997145271914118765555572572572572569762049337429416998714426181310765444437257257257256976204933742941689870432518129754433337257257257256976204933742941689769422417118643322227257257257256976204933742941689769422416118543222117257257257256976204933742941689769422416117543221117257257257256976204933742941689769422416117533211117257257257256976204933742941689769422416117533211117257257257256976204933742941689769422416117533211117257257257256976204933742941689769422416117532211117257257257256976204933742941689769422416117532211117257257257256976204933742941689769422416117532211117257257257256976204933742941689769422416117532211117257257257256976204933742941689769422416117532211117257257257256976204933742941689769422416117532211117257257257256976204933742941689769422416117532211117257257257256976204933742941689769422416117532211117257257257256976204933742941689769422416117532211117257257257256976204933742941689769422416117532211110.0

1.01.21.5

2.02.5

3.03.5

4.0

5.0

6.0

7.0

8.010.012.014.017.020.025.030.035.040.050.070.0100.00.0

1.0

1.2

1.5

2.0

2.5

3.0

3.5

4.0

5.0

6.0

7.0

8.0

10.0

12.0

14.0

17.0

20.0

25.0

30.0

35.0

40.0

50.0

70.0

100.0

Table 8:Muon-electron/photon trigger rate in 100Hz units at

muon p t (GeV) cut

e l e c t r o n E t (G e V ) c u t

725

697

620

493

374

294

168

70770770770768160748536929116696684123161075322111166266266266263756745234727415691653922159643221111539539539539520465372286226133805936201396432111114024024024023913542902301831066750311711753221111127027027027026023118214211164433322118532111183183183183177156119896939282316853211113313313313313112095735731221813642118686868685775945362014128421114444444443403327231411963211292929292926201513975421124242424242116111075431113131313121198754321666665544322133333333221111111111110.0

1.01.21.5

2.02.5

3.03.5

4.0

5.0

6.0

7.0

8.010.012.014.017.020.025.030.035.040.050.070.0100.00.0

1.0

1.2

1.5

2.0

2.5

3.0

3.5

4.0

5.0

6.0

7.0

8.0

10.0

12.0

14.0

17.0

20.0

25.0

30.0

35.0

40.0

50.0

70.0

100.0

Table 9:Muon-beauty electron trigger rate in 100Hz units at

muon p t (GeV) cut

e l e c t r o n E t (G e V ) c u t

725

697

620

493

374

294

168

662662662662639573462356282161936640221596432211115965965965965745144173292631518964392214964322111147847847847846241533726621212474553419128643211111357357357357347313256203160935945281510642211221221221221213188151119955437281896421111391391391391341189271583324201464211939393939184695546261916115321535353535247393127151196311252525252524222017108752113131313131211109654319999988775432177777766543321333333332221122222222211110.0

1.01.21.5

2.02.5

3.03.5

4.0

5.0

6.0

7.0

8.010.012.014.017.020.025.030.035.040.050.070.0100.00.0

1.0

1.2

1.5

2.0

2.5

3.0

3.5

4.0

5.0

6.0

7.0

8.0

10.0

12.0

14.0

17.0

20.0

25.0

30.0

35.0

40.0

50.0

70.0

100.0

Table 10:Muon-jet trigger rate in 100Hz units at

muon p t (GeV) cut

j e t E t (G e V ) c u t

725

697

620

493

374

294

168

180180180180175158127102845237292011854221114343434342393429261713118543211202020202018171513975422111111111110109885432111777776655322111444444333221113333332221111222222211111111111111111111111111111110.0

5.010.015.020.025.030.035.040.045.050.055.060.065.070.075.080.085.090.095.0100.0105.0110.0115.0120.00.0

1.0

1.2

1.5

2.0

2.5

3.0

3.5

4.0

5.0

6.0

7.0

8.0

10.0

12.0

14.0

17.0

20.0

25.0

30.0

35.0

40.0

50.0

70.0

100.0

Table 11:Muon-trigger rate in 100Hz units at

muon p t (GeV) cut

m i s s i n g E t (G e V ) c u t

725

697

620

493

374

294

168

68686868676252433826191611643211114141414131211985432111555554443221112222221111111111110.0

10.020.030.040.050.060.070.080.090.0100.0110.0120.0130.0140.0150.0160.0170.0180.0190.0200.0210.0220.0230.0240.00.0

1.0

1.2

1.5

2.0

2.5

3.0

3.5

4.0

5.0

6.0

7.0

8.0

10.0

12.0

14.0

17.0

20.0

25.0

30.0

35.0

40.0

50.0

70.0

100.0

Table 12:Muon-trigger rate in 100Hz units at

muon p t (GeV) cut

t o t a l E t (G e V ) c u t

725

697

620

493

374

294

168

193193193193188172143111905435271897532111156565656555045403421161310532211252525252423211917119753211131313131313111097654211666666554322111555554433221113333333221111222222221111222221111111111111111111111111111110.0

20.040.060.080.0100.0120.0140.0160.0180.0200.0220.0240.0260.0280.0300.0320.0340.0360.0380.0400.0420.0440.0460.0480.00.0

1.0

1.2

1.5

2.0

2.5

3.0

3.5

4.0

5.0

6.0

7.0

8.0

10.0

12.0

14.0

17.0

20.0

25.0

30.0

35.0

40.0

50.0

70.0

100.0

In the case of two-object triggers at low luminosity we can afford the lowest possible muon cut.It is determined by the muon energy loss in the calorimeters and therefore it varies with .In the barrel it is 4GeV .In the endcaps

it decreases down to

2GeV at

.More precisely one can de?ne the threshold as:GeV

for GeV for GeV for

At high luminosity it is convenient to set the muon threshold for two-object triggers at 4GeV in the entire range.

Ones we ?xed the muon thresholds we can plot the two-object trigger rates as a function of the threshold on the second object.This is done in Figures 9-13.

electron/photon E t cut (GeV)

μ-e /γ t r i g g e r r a t e (H z )

110101010100.1.1.1.2.2.3.3.4.5.6.7.8.10.12.14.17.

20.25.30.35.40.50.70.100.b-electron E t cut (GeV)

μ-e b t r i g g e r r a t e (H z )

10101010100.1.1.1.2.2.3.3.4.5.6.7.8.10.12.14.17.20.25.30.35.40.50.70.100.Figure 9:

Muon-electron/photon trigger rate for 2-4GeV at .

Figure 10:Muon-beauty electron trigger rate for

2-4GeV at .

jet E t cut (GeV)

μ-j e t t r i g g e r r a t e (H z )

0.5.10.15.20.25.30.35.40.45.50.55.60.65.70.75.80.85.90.95.100.105.110.115.120.Figure 11:Muon-jet trigger rate for

2-4GeV at

E t miss cut (GeV)

μ-E t m i s s t r i g g e r r a t e (H z )

1010

0.10.20.30.40.50.60.70.80.90.100.110.120.130.140.150.160.170.180.190.200.210.220.230.240.ΣE t cut (GeV)

μ-ΣE t t r i g g e r r a t e (H z )

0.20.40.60.80.100.120.140.160.180.200.220.240.260.280.300.320.340.360.380.400.420.440.460.480.Figure 12:Muon-trigger rate for

2-4GeV at

Figure 13:Muon-trigger rate for

2-4GeV at

All the results are summarised in Tables 13.Here we have chosen the threshold to keep the total trigger rate at 30kHz.The rates of calorimeter triggers are taken from [10].The case of DAQ limited to 75kHz,i.e.the total trigger rate of 25kHz,is presented in Table 14.

3Conclusions

Presented muon/calorimeter trigger study accomplishes the task of calculating the 1st level trigger output rates.A set of trigger thresholds is proposed to maintain the 1st level output rate at 30kHz.The thresholds are low enough to ful?l the CMS physics program presented in the CMS Technical Proposal [1].

The results indicate a possibility of important gain in several physics channels,especially in CP violation studies:

CP violation —angle :B J/K ;J/

or ee;b

or e

-e trigger —

2-4GeV;

4GeV

CP violation —angle :B

;b or e

-jet trigger,where the

pair is treated as a “jet”—

2-4GeV;

10GeV

CP violation —oscillations:B D KK

;

b or e -jet trigger,where the pair is treated as a “jet”

—

2-4GeV;

10GeV

Presented trigger thresholds should not be blindly used in physics simulation.They should rather be considered as examples chosen to demonstrate trigger capabilities.One should also keep in mind that the results were obtained with fast simulation program and they should be checked later by more detailed (full GEANT/CMSIM )simulations.

Table13:Trigger rates for selected cuts.LV2input rate=100kHz.

rate(kHz)thresholds

individual individual

e2-4748

e2-44

j2-410440

2-440460

2-41004250

Table14:Trigger rates for selected cuts.LV2input rate=75kHz.

trigger thresholds

individual

2-4

References

[1]The Compact Muon Solenoid—Technical Proposal,CERN/LHCC94-38.

[2]CMS TN/94-219,Ph.Busson,J.Varela,Calorimeter Trigger in CMS Algorithm studies I.

[3]CMS TN/95-027,C.Lourenco,J.Varela,A?ne granularity calorimeter trigger for CMS.

[4]CMS TN/95-143,J.Varela,Requirements for a?ne grained calorimeter trigger,

presented at the First Workshop on Electronics for LHC Experiments,Lisbon,11-15September1995.

[5]CMS TN/95-197,C.Lourenco,A.Nikitenko,J.Varela,

A low1st level single electron trigger for beauty studies in CMS.

[6]CMS TN/96-021,R.Nobrega,J.Varela,

The CMS electron/photon trigger:simulation study with CMSIM data.

[7]CMS TN/94-285,S.Dasu,https://www.360docs.net/doc/3013313486.html,ckey,W.H.Smith,W.Temple,

CMS Level1Calorimeter Trigger Performance Studies.

[8]CMS TN/95-111,S.Dasu,https://www.360docs.net/doc/3013313486.html,ckey,W.H.Smith,W.Temple,

CMS Missing Transverse Energy Trigger Studies.

[9]CMS TN/95-112,S.Dasu,https://www.360docs.net/doc/3013313486.html,ckey,W.H.Smith,W.Temple,

New Algorithms for CMS Electron/Photon Trigger—Use of Fine Grain Calorimeter Data.

[10]CMS TN/95-183,S.Dasu,https://www.360docs.net/doc/3013313486.html,ckey,W.H.Smith,W.Temple,

CMS Level1Calorimeter Trigger Performance on Technical Proposal Physics.

[11]CMS TN/96-066,S.Dasu,W.H.Smith,CMS Level-1Jet Trigger Study.

[12]S.Dasu,C.Lourenco,A.Nikitenko,R.Nobrega,W.H.Smith,J.Varela,

Simulation of the CMS Level-1Calorimeter Trigger,presented at the VI International Conference on Calorime-try in High Energy Physics,Frascati,Italy,June1996.

[13]CMS TN/92-039,M.Konecki,J.Kr′o likowski,G.Wrochna,

Simulation study of the RPC based,single muon trigger for CMS.

[14]CMS TN/94-138,M.Huhtinen,G.Wrochna,

Estimation of the RPC muon trigger rates due to neutral particles.

[15]CMS TN/94-282,A.Csilling,M.Konecki,J.Kr′o likowski,G.Wrochna,Muon Trigger Rates.

[16]CMS TN/95-150,M.′Cwiok,G.Wrochna,

Muon rates in UA1,D0,CDF and CMS predicted by PYTHIA5.7.

[17]A.Fengler,Double muon trigger rates in CMS experiment,PHD thesis,Warsaw University,1996,

http://cmsdoc.cern.ch/fengler/MSc/trig2mu.ps.Z

[18]J.Pliszka,G.Wrochna,Fast simulation tools for trigger study,talk given at the CMS Software Week,CERN,

October1996,http://cmsdoc.cern.ch/wrochna/GW